Promoting Marriage Has Failed and Is Unnecessary to Cut Poverty

The Brookings/AEI anti-poverty consensus consists of three points: education, work, and marriage. In a prior post, I attacked the education focus as misguided: we've been massively ramping up educational attainment for decades with no poverty gains to show for it. Also in a prior post, I attacked the work focus as similarly misguided: most poor people cannot work more because they are children, elderly, disabled, students, or already fully employed, among other problems. In today's post, I take on the marriage point.

1. Marriage Promotion Doesn't Work

Marriage promotion policy has been an utter failure, and not for lack of trying.

One of the main points of so-called Welfare Reform was to end the scourge of single motherhood and promote marriage, "the foundation of a successful society." Since that reform, which massively spiked extreme poverty in the US, the rate of birth to unwed mothers has continued to go up, and marriage rates have continued to go down. Nothing has been able to reverse this trend.

As part of Welfare Reform and other related measures, the government also got directly in the business of promoting marriage via projects like Building Strong Families, Supporting Healthy Marriage Project, and the Healthy Marriage Initiative more generally. As Bryce Covert has extensively pointed out at The Nation and elsewhere, assessments of these programs have found them to be utter failures. Nonetheless, states still redirect TANF funds meant to provide cash assistance to poor families to these dead-end ideology projects.

The Brookings/AEI report, which purports to be an evidence-based inquiry into how best to cut poverty, actually admits that efforts to inch the marriage rate up have been a failure, but nonetheless insists that we embark upon a vague cultural campaign to promote it:

So what can be done? We’ve said that marriage matters. But past government efforts to encourage unmarried parents to marry have not proven very effective. Promoting marriage to strengthen American families isn’t primarily an issue of specific policies or programs in any case: it’s in large part a question of culture. Political leaders, educators, and civic leaders—from both the political left and right—need to be clear and direct about how hard it is to raise children without a committed co-parent.

This is not a real policy in any meaningful sense. How would we get "political leaders, educators, and civil leaders" to promote these messages and where? Are they going to get funding to put up "Marriage Is Cool" billboards in their communities? Who will oversee this vague marriage cultural campaign? This is the kind of proposal you get from people who still cling on to something they know is a failure. It's intellectualized wishing for cultural transformations, not policy.

2. Marriage Is Not Necessary to Dramatically Cut Poverty

Typically, what people focus on with marriage arguments is children. And the Brookings/AEI report is no exception in this regard. Despite their pretention of exhaustively reviewing the evidence regarding how you might cut child poverty, they somehow missed the part where other countries do so very effectively, despite having similar rates of single motherhood and unwed births as we do.

As I discussed last year, cross-country research shows that differences in family composition across countries do not explain differences in childhood poverty across countries. The countries with the lowest child poverty rates in the world have similar rates of single motherhood as the US (this using 2000 LIS data):

Yet they manage to keep child poverty quite low, overall and in single-mother families, by having a nice welfare state (T&T refers to taxes and transfers):

As much as Brookings/AEI touts the wonders of married-parent families in the US, the reality is that they too have mind-bogglingly high levels of poverty, compared to low-poverty countries. In the same 2000 LIS data used above, child poverty in married-parent families was 6x higher in the US than in Sweden and 7.3x higher in the US than in Finland:

In fact, US married-parent child poverty is so high that it's higher than the child poverty rate in single-mother families in the low-poverty Nordics:

If you want to keep child poverty low (in all family types, not just unmarried families), one proven way to do that is to provide good child benefits. Sitting at the bottom of the OECD world in social expenditures on child-related family benefits helps to explain far more of our child poverty problem than declining marriage does:

If we beefed these benefits up, we would see child poverty (and other poverty) fall dramatically. For instance, as I discussed earlier this year, if we provided each family $300/month for every child they are caring for, we could cut child poverty by 40% and overall poverty by 23%. Even after doing that, we would still only find ourselves in the middle of the child-benefit pack for OECD countries (meaning there is even more room for spending on child care and related benefits, if we want them).

3. Why Not Shake Up Families In Other Ways?

The fact that marriage promotion doesn't work and you can easily cut child poverty without is all you need to know about this focus. Things that don't work and aren't necessary are things you discard as useless unless you're an ideologue.

Nonetheless, the marriage obsession is interesting for at least one reason. That reason is this: if reorganizing people into different households is an appropriate aim of anti-poverty policy, why on earth would you stop at getting low-income people to marry one another?

You see, the way this marriage anti-poverty stuff is supposed to work is pretty basic. For poverty calculation purposes, we draw circles around groups of individuals, placing all of them into different pods (household or family depending on how you want to measure). For an unnecessary graphic of what I mean, here is this (red dots are individuals):

After we place each person in a pod, we go to each pod, add up all the income in that pod, and see if that combined income is enough to bring the overall pod out of poverty. The more people in the pod, the more combined income the pod needs to be out of poverty.

The primary way marriage is supposed to cut poverty is by reshuffling who is in which pod. So, imagine that the one-person pod on the top left contains a guy who makes $15,000 per year and the three-person pod on the top right contains a mother and two kids whose total income is $12,000 per year. The three-person pod is in poverty as $12,000 is not enough for a family of three. But if we move the guy from the one-person pod into the three-person pod (and bring his income with him), then we get a new four-person pod with an income of $27,000, which is indeed enough for the pod to be out of poverty. Thus, through this reconfiguration of pods, we pulled three people out of poverty.

Once you understand how this marriage anti-poverty story works, two things should immediately occur to you. First, it's not marriage that's cutting poverty here. It's the recombination of income. If you marry someone who does not add extra income to your family, that will do nothing to reduce your chance of poverty and, because your family is now bigger, it could even drag your family into poverty. Adding someone to a resource pod only reduces that pod's chance of poverty where, as a result of the addition, the increase in the pod's income is greater than the increase in the pod's poverty line. Sometimes marriages accomplish this, sometimes they do not.

Second, you can recombine income across pods in much more effective ways than convincing two low-income people to marry and move in together. For instance, why not just move some of the high-earners from the very rich pods into the very poor pods? Alternatively, why not move some of the people from the very poor pods into the very rich pods? This kind of reshuffling of pod membership would seem to directly tackle the poverty problem in ways that getting two poor people to marry (note nobody ever seems to advocate imploring rich people to marry poor people, only exhorting poor people to marry each other) does not. If reshuffling pod membership (which is all marriage anti-poverty stuff is about) is fair game for poverty reduction, then these cross-class reshufflings offer a far more promising strategy.

The only objection I can see to class-based reshuffling is that you can't force people to move into families and households that they don't want to be in, and there is no reason to think that we could do anything policy-wise that would get people to take up this reshuffling themselves. That's a fair enough objection. In fact, that's the marriage promotion objection.